Emerging Role of Social Media Analytics in Health Care and BioMedical Research


Published on

Keynote talk by Dr. Sudha Ram at the BIG Data SUMMIT, July 31, 2012.

Published in: Technology
1 Like
  • Be the first to comment

No Downloads
Total views
On SlideShare
From Embeds
Number of Embeds
Embeds 0
No embeds

No notes for slide

Emerging Role of Social Media Analytics in Health Care and BioMedical Research

  1. 1. The Emerging Role of Social MediaAnalytics in Health Care and Medicine Dr. Sudha Ram Director, INSITE Center for Business Intelligence and Analytics McClelland Professor of MIS and Computer Science Member, BIO5 and IE Eller College of Management University of Arizona Tucson, AZ 85721 Email: ram@eller.arizona.edu INSITE: www.insiteua.org July 31, 2012
  2. 2. Web: Limitless Possibilities• Treasure Trove of Big Data, Dynamic Content, Hard Problems• Volume, Velocity, Variety of Data• Unstructured Data, Timestamped, location based• Predictive Analytics and Data Mining to the Rescue
  3. 3. Web – Interaction Networks Model as an Interaction Network
  4. 4. Network of Interactions• Explicit Networks: Followers and Following on Twitter, Friendships on Facebook, LinkedIn Professional NetworkBut….. More interesting….• Implicit Networks: Based on Interactions on forums, FB Pages, Wikipedia, Yelp, Foursquare….• These hidden networks can reveal a lot of interesting patterns. 4
  5. 5. Implicit Interaction Network
  6. 6. Implicit Interactions• Wikipedia: Edits and Collaborations on an article, across articles, on a topic…• Facebook: Posts and comments• Twitter: Tweets, replies, retweets• Blogs: Comments and responses• Forums: Threads, responses 6
  7. 7. Media News Diffusion Networks Timeline – 3 monthsSeptember – November 2011
  8. 8. BBCNetwork
  9. 9. Nytimes
  10. 10. Leveraging Social Media• Health care• Biomedical Research• Personalized Medicine 12
  11. 11. Online Healthcare Communities• Patients Like Me• CureTogether• MedHelp• 23andMe 13
  12. 12. OHCs• Individuals sharing Information• Diseases, Treatments, Symptoms, disease progression, blood tests and results, demographics,• Large Populations of cases and controls• Rapidly producing clinical research results• PLM – 150,000+ individuals, MedHelp: 12 million+ users 14
  13. 13. OHC Examples• ALS (amyotrophic lateral sclerosis) and Lithium Treatment (PLM, 2011)• Parkinson’s Disease – 2 novel genes Genotyping services, phenotyping information gathered online (23andMe, 2011)• Multifocal lens implant failures (MedHelp) 15
  14. 14. Other examples• MyHeartMap: Crowdsourcing to located AEDs in Philadelphia• RxAnte: Medication Adherence and Intervention Design• Genomera: Socially designed health care studies 16
  15. 15. Benefits of Using Social Media• Accelerate the pace of research• Recruitment of subjects for clinical trials• Replicate GWAS quickly and reliably• Answer Community generated questions• Live experiments 17
  16. 16. INSITE Center• Collect data, structure and clean it to create datasets• Build tractable models and algorithms for detecting patterns and further our understanding• Observe structures not visible on a smaller scale• Live Analytics and Real Time Business Intelligence
  17. 17. Conclusion• Free the Data• Link Social Media generated Data with Transactional data• Discovery Driven Approach 19
  18. 18. Discussion Thank you! Questions/Comments??ram@eller.arizona.edu 20